Major health disparities exist in obesity and diabetes among ethnic groups in the US. Mexican American (MA) women are particularly afflicted by prediabetes as 45% of MA women are obese and have an age-adjusted 9.3% diabetes rate, compared to 33% obesity and 5.4% diabetes rates in Caucasian women. Pancreatic ?-cell dysfunction is a major factor in the conversion of pre-diabetes to diabetes. By the time type 2 diabetes is diagnosed, 80% of ?-cell function is lost; while 50% is lost by the time prediabetes develops. Thus, early intervention at least by the prediabetes stage is critical. Strong evidence suggests that glucagon-like peptide-1 (GLP-1) preserves ?-cell function. Our goal is to determine whether intense lifestyle change with and without a GLP-1 analog improves ?-cell function in MA women with pre-diabetes, whether the effect is maintained after drug discontinuation, and whether genetic variations predict this response. Predictive genes are likely to be identified because most genes identified for diabetes-related traits influence ?-cell function. Improvement in ?-cell function can potentially prevent the downward spiral of metabolic disease in MA plagued by poverty, limited education, and health disparities. 1. We hypothesize that lifestyle change plus a GLP-1 analog improves ?-cell function in obese individuals with pre-diabetes more than lifestyle change alone. We will compare the effects of a meal replacement (MR) and culturally sensitive lifestyle modification program with and without liraglutide. The primary endpoint will be improved ?-cell function, measured by the acute insulin response, AIRg, determined from the frequently sampled intravenous glucose tolerance test, FSIGT). DI (a measure of AIR and insulin sensitivity) will also be calculated. Secondary endpoints include changes in ?-cell markers (C-peptide, insulin genic index (IGI), glucagon); novel plasma biomarkers that predict diabetes and inflammation; and metabolic syndrome components. 2. We hypothesize that genetic variations that are associated with ?-cell dysfunction will predict the ?-cell response to intervention. We will use a selected group of SNPs on the Metabochip to correlate polymorphisms of T2DM-related genes implicated in ?-cell dysfunction with insulin responses and other phenotypes in liraglutide-treated subjects from Aim 1. This investigation targets serious health disparities in metabolic disease in a highly vulnerable, rapidly growing population, testing novel culturally-focused intervention strategies and identifying genetic biomarkers of ?-cell response to a pharmacologic intervention that targets the pancreatic ?-cell. These data ultimately set the stage for an intervention trial to prevent diabetes, a major chronic and costly disease, in MA, which could substantially improve the health disparity associated with obesity and diabetes.